Multimodality assessment of sleep outcomes in people living with HIV performed using validated sleep questionnaires

Abstract

Studies conducted in people living with HIV (PLHIV) report high rates of sleep disturbance, without a clear explanation as to cause or effect. Therefore, we proposed use of multiple validated questionnaires that would allow a more comprehensive evaluation of sleep quality in PLHIV. We administered eight validated sleep and wellbeing questionnaires, recording different aspects of sleep in order to provide a comprehensive description of sleep quality, quantity, daytime functioning, wakefulness, and general wellbeing. Associations with demographics and clinical data were analyzed by univariable/multivariable analyses. Of 254 subjects 99% were male (98% men who have sex with men), 88% white, mean age 41 (SD ± 9.9) years, HIV duration eight years (SD ± 6.3), 94% were on antiretroviral therapy, mean CD4 cell count was 724 cells/mm³, 81% had HIV RNA<40 copies/ml, 72% were university educated, and 60% used ‘chemsex’ drugs. Almost half (45%) reported poor sleep quality, 22% insomnia, 21% daytime sleepiness, and 33% fatigue. As individual factors, HIV duration ≥10 years, anxiety, depression, and recreational drug use were associated with poor quality sleep, fatigue, and poorer functional outcomes (p ≤ 0.05). The prevalence of sleep disturbance was high in our cohort of PLHIV. Sleep disturbance was associated with longer duration of HIV infection, depression, anxiety, and recreational drug use.

Keywords

Sleep quality HIV wellbeing

Background

A strong correlation has been observed between sleep disorders and poor health outcomes, and a high prevalence of sleep disturbance has been described in people living with HIV (PLHIV) (30–73%^1–4 versus 10–20% in the general population^5,6). Impaired sleep quality in PLHIV may negatively impact immune function and pain threshold, and lead to functional disability as well as contributing to development of co-morbidities, especially related to mental and cardiovascular health.^7–10

Sleep disorders in PLHIV have been extensively studied; however, the influence and bidirectional causality of age, choice of antiretroviral drugs, co-medications, recreational drug use, drug–drug interactions, and lifestyle factors on sleep-related outcomes remain unclear.

The objective of our study was to describe sleep outcomes in PLHIV by performing a detailed comparative analysis of different validated sleep questionnaires, the domains they represent (herein called multimodality assessment), and to determine associations with sleep outcomes.

Methods

People with confirmed HIV diagnosis, 18 years or older, who attended a single central London HIV outpatient clinic (56 Dean Street) between March 2016 and May 2017 were approached during regular clinic visits and offered study participation (DS HIV Cohort Study). All participants provided written informed consent. Ethics approval was obtained from the London Dulwich Research Ethics Committee and Health Research Authority (IRAS 165402).

Following enrolment, participants completed eight validated, self-reported sleep and wellbeing questionnaires: (1) Functional Outcomes of Sleep Questionnaire (FOSQ): 30-item tool to measure the effect of sleep disorders on multiple activities of daily living including activity level, vigilance, intimacy, general productivity, and social outcomes (range 5–20)¹¹; (2) Pittsburgh Sleep Quality Index (PSQI): 19-item tool measuring overall sleep quality across seven domains (range 0–21)¹²; (3) Insomnia Severity Index (ISI): 7-item instrument measuring severity of insomnia symptoms (range 0–28; >15 = abnormal)¹³; (4) Epworth Sleepiness Scale (ESS): sum of eight daytime situations where dozing may occur (range 0–24; ≥10 = excessive daytime sleepiness)¹⁴; (5) Fatigue Severity Scale (FSS): 9-item tool to assess the severity of fatigue (range 9–63; >36 = abnormal)¹⁵; (6) Generalized Anxiety Disorder (GAD): sum of seven measures to assess anxiety (range 0–21)¹⁶; (7) Patient Health Questionnaire (PHQ): sum of nine measures to assess depression (range 0–27)¹⁷; and (8) Wellness Thermometer (Wellness): a single question evaluating patient-perceived wellness from 1 (not feeling well) to 10 (feeling well).¹⁸ For all questionnaires, a higher score indicates a poorer performance, except FOSQ and Wellness in which a higher score reflects a better status.

Participants had the option to complete the questionnaires at their clinic visit, or at home and return via post. Additional data were collected using a case report form, including socio-demographic characteristics, HIV-related data, current/previous treatments, recreational drug and alcohol use, medical history (including sexually transmitted infections [STIs]), and laboratory values.

Data analysis

Descriptive statistics were used to define demographics, lifestyle, and HIV-related characteristics and to characterize the questionnaire scores in the cohort. Where available, questionnaire results were categorized according to standard criteria for each scale (not available for FOSQ and Wellness). Normality of scores was tested using the Shapiro–Wilk test and visual inspection of histograms. Based on these results, Spearman’s rho correlations were used to examine the relationship between instruments. Internal consistency was measured with the Cronbach alpha (α).

Factors associated with sleep and wellness outcomes were identified using univariable and multivariable regression. Linear regression was used for continuous outcomes (FOSQ and Wellness), reporting coefficients (β) with 95% confidence intervals (CIs). Factors associated with poor sleep quality (PSQI >5), insomnia (ISI ≥15), sleepiness (ESS ≥10), fatigue (FSS ≥36), anxiety (GAD ≥10), and depression (PHQ ≥10) were analyzed using logistic regression, reporting odds ratios (ORs) with 95% CIs. Given the high correlations observed between questionnaires, multivariable models for (i) all sleep outcomes and (ii) all wellbeing outcomes were built using a set of covariates common to each group of outcomes. Covariates associated with at least one score (p < 0.1) under univariable analysis were included in all multivariable models for that group (i.e. sleep/wellness). Covariates assessed were age, nationality, highest education level obtained, alcohol consumption (>once/week), recreational drug use (specific drug use in previous six months), STIs in previous six months, time since HIV diagnosis, current CD4 cell count, and HIV viral load (VL). Laboratory results were the most recent within six months. Anxiety and depression (GAD and PHQ) were entered as dichotomous covariates to sleep models and a sleep quality indicator (PSQI >5) was entered in wellbeing models. All covariates were coded as binary or categorical. Collinearity among covariates was examined using variance inflation factors; results did not indicate any multicollinearity. Observations with missing data were dropped from the models. Stata (version 14.2; StataCorp) was used for analysis and the level of significance set at p < 0.05.

Table 2.

Independent multivariable regression models to determine predictors of sleep and wellness outcomes (univariable analyses are not shown).

	Functional outcomes (FOSQ, continuous)	Sleep quality (PSQI >5)	Insomnia (ISI ≥15)	Sleepiness (ESS ≥10)	Fatigue (FSS ≥36)	Depression (PHQ ≥10)	Anxiety (GAD ≥10)	Wellness (continuous)
	β (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	β (95%CI)
Age, category (years)
<40	ref	ref	ref	ref	ref	ref	ref	ref
40–49	−0.2 [−1.1,0.6]	0.6 [0.3,1.5]	1.0 [0.3,3.3]	1.1 [0.4,2.8]	0.8 [0.3,2.0]	1.2 [0.4,3.2]	0.6 [0.1,2.1]	−0.5 [−1.1,0.1]
≥50	0.1 [−0.9,1.1]	1.4 [0.5,4.1]	2.9 [0.6,13.3]	1.4 [0.4,4.7]	2.8+ [0.9,8.2]	0.5 [0.1,2.1]	0.2* [0.0,1.0]	0.7 [−0.1,1.4]
Highest level of education attained
Secondary	ref	ref	ref	ref	ref	ref	ref	ref
College/vocational	3.4*** [1.8,5.1]	1.1 [0.2,5.2]	1.1 [0.1,10.3]	2.4 [0.3,17.8]	0.5 [0.1,2.7]	2.2 [0.3,15.4]	0.7 [0.1,5.1]	0.1 [−1.1,1.3]
University	2.7*** [1.2,4.2]	0.9 [0.2,3.7]	1.4 [0.2,11.4]	1.4 [0.2,9.6]	0.7 [0.1,3.1]	1.6 [0.3,10.1]	0.3 [0.1,2.1]	0.7 [−0.5,1.8]
Employed	0.6 [−0.5,1.7]	1.3 [0.4,3.9]	2.7 [0.4,19.4]	1.9 [0.5,7.7]	1.5 [0.5,4.8]	0.7 [0.2,2.4]	0.4 [0.1,1.8]	1.0* [0.2,1.8]
Alcohol, >once/week	0.1 [−0.7,0.8]	0.5+ [0.2,1.1]	0.7 [0.2,2.2]	2.3+ [0.9,5.5]	1.8 [0.8,4.1]	0.4+ [0.2,1.1]	1.4 [0.5,4.1]	0.5+ [−0.1,1.1]
STIs in past six months						1.5 [0.6,4.3]	3.2+ [1.0,10.7]	−0.5 [−1.2,0.1]
Recreational drugs in last six months
Crystal meth	−0.2 [−1.2,0.8]	0.5 [0.2,1.4]	0.4 [0.1,1.5]	0.3+ [0.1,1.1]	1.9 [0.7,5.3]	4.2* [1.1,15.6]	3.2 [0.7,15.7]	−0.4 [−1.3,0.4]
Ketamine	−0.3 [−1.8,1.2]	1.8 [0.3,10.0]	0.6 [0.1,6.4]	1.1 [0.2,8.0]	2.1 [0.4,10.2]	1.1 [0.3,4.7]	0.7 [0.1,4.1]	−0.2 [−1.3,0.9]
Cocaine	−0.2 [−1.1,0.7]	1.4 [0.5,3.8]	1.8 [0.4,7.0]	1.5 [0.4,5.0]	0.6 [0.2,1.8]
Cannabis	0.1 [−0.8,1.0]	1.1 [0.4,2.7]	1.9 [0.5,6.4]	0.3+ [0.1,1.1]	0.8 [0.3,1.9]	1.2 [0.5,3.3]	1.2 [0.4,3.5]	−0.7* [−1.4,−0.0]
Amyl nitrate						1.8 [0.7,4.7]	1.7 [0.6,5.3]	−0.1 [−0.7,0.5]
G (GBL, GHB)						0.6 [0.2,2.1]	0.3+ [0.1,1.3]	0.7+ [−0.1,1.5]
Time since HIV diagnosis
<5 yrs	ref	ref	ref	ref	ref
5–9 yrs	−0.4 [−1.3,0.5]	1.1 [0.4,2.8]	0.9 [0.2,3.8]	2.1 [0.7,6.2]	1.1 [0.4,2.9]
≥10 yrs	−1.2* [−2.1,−0.2]	3.0* [1.1,7.9]	2.5 [0.7,9.3]	2.4 [0.8,7.1]	2.9* [1.1,7.7]
ART category
2NRTI+1NNRTI (no EFV)	ref	ref	ref	ref	ref	ref	ref	ref
2NRTI+1NNRTI (EFV)	−0.7 [−1.9,0.5]	3.2+ [0.9,10.7]	3.9 [0.6,25.2]	1.4 [0.4,5.3]	1.2 [0.4,3.8]	0.4 [0.1,2.1]	0.3 [0.1,2.0]	1.0* [0.1,2.0]
2NRTI+1PI	0.5 [−0.8,1.8]	2.0 [0.6,6.9]	1.3 [0.1,11.9]	0.6 [0.1,3.0]	0.3+ [0.1,1.0]	0.9 [0.2,4.7]	0.5 [0.1,3.2]	0.9+ [−0.1,1.9]
2NRTI+1INSTI	−0.3 [−1.5,0.9]	2.8+ [0.8,9.2]	3.0 [0.4,19.9]	1.1 [0.3,4.1]	0.8 [0.2,2.6]	1.7 [0.4,6.9]	0.5 [0.1,2.4]	0.5 [−0.4,1.4]
Other	0.6 [−1.2,2.3]	0.2 [0.0,2.2]		1.4 [0.2,11.0]	0.1 [0.0,1.5]	1.0 [0.1,13.4]	1.3 [0.1,24.6]	0.5 [−0.9,1.9]
No ART	−0.2 [−2.0,1.6]	0.5 [0.1,4.1]	3.3 [0.2,50.0]	1.3 [0.1,12.1]	0.7 [0.1,4.8]	2.7 [0.3,21.8]	2.9 [0.4,24.4]	1.0 [−0.4,2.4]
Generalized Anxiety Disorder
None	ref	ref	ref	ref	ref
Mild	−0.3 [−1.3,0.6]	1.6 [0.7,4.1]	1.6 [0.4,6.2]	2.4 [0.8,7.1]	1.0 [0.4,2.8]
Moderate/severe	−0.6 [−2.1,0.8]	5.8* [1.2,27.0]	1.2 [0.2,7.5]	4.4* [1.0,18.5]	1.8 [0.5,6.6]
Patient Health Questionnaire (depression)
None	ref	ref	ref	ref	ref
Mild	−1.9*** [−2.9,−1.0]	3.3* [1.3,8.4]	4.7* [1.0,21.8]	1.7 [0.5,5.4]	8.2*** [2.8,23.5]
Moderate/severe	−2.5*** [−3.8,−1.2]	7.9** [2.2,28.4]	16.1** [2.6,100.1]	4.6* [1.2,18.0]	18.4*** [4.9,69.0]
Poor sleep quality (PSQI >5)						8.3*** [3.1,22.2]	13.2*** [3.6,48.0]	−1.4*** [−1.9,−0.8]
Observations (% of total)	186 (73)	197 (78)	137 (54)	202 (80)	204 (80)	205 (81)	204 (80)	207 (81)

ART: antiretroviral therapy; EFV: efavirenz; ESS: Epworth Sleepiness Scale; FOSQ: Functional Outcomes of Sleep Questionnaire; FSS: Fatigue Severity Scale; GAD: Generalized Anxiety Disorder; GBL: γ-butyrolactone; GHB: γ-hydroxybutyrate; INSTI: integrase strand transfer inhibitor; ISI: Insomnia Severity Index; NNRTI: non-nucleoside reverse transcriptase inhibitor; NRTI: nucleoside reverse transcriptase inhibitor; PHQ: Patient Health Questionnaire; PSQI: Pittsburgh Sleep Quality Index; STI: sexually transmitted infection.

Each column represents an independent regression model. Point estimates are odds ratios (OR) derived from logistic regression for dichotomous outcomes and coefficients from linear regression (β) for continuous outcomes with the corresponding 95% confidence interval. For the continuous outcomes (FOSQ and WT), a lower score indicated a poorer performance.

Simple regression models were run to determine covariates significantly associate with each outcome at the p < 0.1 level. In the multivariable models, all covariates found to be associated with any sleep outcome were entered into the sleep outcome models; all outcomes associated with any mental health/wellness outcome were entered into the models for these outcomes (PHQ, GAD, Wellness).

+p < 0.1, *p < .05, **p < .01, ***p < .001.

Results

We recruited 254 PLHIV, of whom >99% were male (98% men who have sex with men [MSM]), 88% white, mean age 41 (SD ±9.9) years, and mean time since HIV diagnosis of eight years (SD ±6.3) (Table 1). Ninety-four percent were on antiretroviral therapy (ART), 24% on efavirenz (EFV) and 33% on integrase strand transfer inhibitors (INSTIs). HIV VL was <40 copies/ml in 81%, and mean CD4 cell count 724 cells/mm³ (SD ±493) at the time questionnaires were completed. Eighty-seven percent of participants were employed, 72% university educated, 25% smokers, 42% used alcohol more than once a week, 61% reported recreational drug use (60% reported use of mephedrone, crystal methamphetamine and γ-hydroxybutyrate/γ-butyrolactone to enable, enhance or prolong sexual interactions [chemsex]), and 18% used recreational drugs intravenously. One-third (31%) had been diagnosed with an STI in the previous six months.

Table 1.

Sleep and quality of life characteristics.

Characteristics	Total (n = 254)
Patient characteristics
Age (years)	40.8 (9.9)
Male	253 (99.6%)
MSM	250 (98.4%)
White ethnicity	221 (87.7%)
British origin	147 (57.9%)
Employed	219 (86.9%)
University educated	183 (72.0%)
Tobacco smoking (current)	61 (24.6%)
Alcohol drinking (>once/week)	105 (41.7%)
Recreational drug use in past six months	150 (60.7%)
Crystal methamphetamine	52 (20.7%)
Mephedrone	72 (29.3%)
Cocaine	59 (24.0%)
Ketamine	20 (8.2%)
GBL, GHB	67 (27.1%)
MDMA	43 (17.4%)
Amyl nitrate	81 (33.1%)
Cannabis	58 (23.4%)
STIs within previous six months	77 (30.6%)
HIV characteristics
Time since HIV diagnosis (years)	8.1 (6.3)
HIV RNA <40 copies/ml	199 (80.9%)
CD4 cell count (cells/mm³)	723.6 (493.2)
ART category
2NRTI+1NNRTI (with EFV)	60 (24.2%)
2NRTI+1NNRTI (without EFV)	37 (14.9%)
2NRTI+1PI	42 (16.9%)
2NRTI+1INSTI	81 (32.7%)
Other	14 (5.6%)
No ART	14 (5.6%)
Sleep characteristics from questionnaires
Functional outcomes of sleep questionnaire (FOSQ; α = 0.86), n = 222
Global score (5–20)	17.0 (2.9)
Pittsburgh Sleep Quality Index (PSQI; α = 0.77), n = 233
Global score (0–21)	5.9 (3.7)
PSQI >5 (poor quality sleep)	104 (44.6%)
Insomnia Severity Index (ISI; α = 0.92), n = 177
Moderate insomnia (15–21)	31 (17.5%)
Severe insomnia (≥22)	8 (4.5%)
Epworth Sleepiness Scale (ESS; α = 0.76), n = 241
Sleepiness (10–17)	45 (18.7%)
Severe sleepiness (≥18)	5 (2.1%)
Fatigue Severity Scale, n = 245
Possible fatigue (≥36)	82 (33.5%)
Mental health and quality of life characteristics
Generalized Anxiety Disorder (GAD; α = 0.90), n = 241
Mild anxiety (5–9)	67 (27.8%)
Moderate anxiety (10–14)	22 (9.1%)
Severe anxiety (≥15)	11 (4.6%)
Patient Health Questionnaire (PHQ; α = 0.88), n = 241
Mild depression (5–9)	75 (31.1%)
Moderate depression (10–14)	24 (10.0%)
Severe depression (≥15)	21 (8.7%)
Wellness thermometer, n = 245
Wellness score (1–10)	6.7 (2.2)

ART: antiretroviral therapy; EFV: efavirenz; GBL: γ-butyrolactone; GHB: γ-hydroxybutyrate; INSTI: integrase strand transfer inhibitor; MDMA: 3,4-methylenedioxy-methamphetamine; MSM: men who have sex with men; NNRTI: non-nucleoside reverse transcriptase inhibitor; NRTI: nucleoside reverse transcriptase inhibitor; PI: protease inhibitor; STI: sexually transmitted infection.

Results are mean (standard deviation) or n (%) where % is calculated from the total non-missing responses (n). For all measures except FOSQ and the Wellness thermometer, a lower score indicates better performance. For FOSQ and Wellness, a higher score indicates better performance.

α is the Cronbach’s alpha for scale reliability (computed using the individual items for ISI, ESS, GAD, and PHQ and the domain results for FOSQ and PSQI). There is no α for the Wellness thermometer as this is a single-item questionnaire.

Individually, questionnaires were well answered with response rate >90% for six of eight questionnaires; 47% completed all questionnaires.

Questionnaire scores are presented in Table 1. Mean FOSQ score was 17.0 (SD ±2.9) indicating good daytime functioning. Overall, 45% reported poor sleep quality (PSQI >5), 22% reported insomnia (ISI ≥15), 21% reported excessive daytime sleepiness (ESS ≥10), and 33% experienced fatigue (FSS ≥36). Moderate to severe anxiety was observed in 14% of the cohort (GAD ≥10); 19% experienced moderate to severe depression (PHQ ≥10) and mean Wellness was 6.7 out of 10 (SD ±2.2).

All questionnaires were strongly correlated (p < 0.001 for each pairwise comparison). The strongest correlations were observed between the sleep quality (PSQI) and insomnia (ISI) questionnaires (rho = 0.79) and the anxiety (GAD) and depression (PHQ) questionnaires (rho = 0.73). Internal consistency was acceptable for all questionnaires (α = 0.76–0.92) (Table 1).

Predictors of sleep and wellbeing outcomes

Functional outcomes (FOSQ)

Better functional outcomes were associated with college education (β = 3.4; 95%CI 1.8, 5.1; p < 0.01) and university education (β = 2.7; 95%CI 1.2, 4.2; p < 0.01) compared to secondary education. Poorer FOSQ outcomes were associated with mild depression (β = −1.9; 95%CI −2.9, −1.0; p < 0.01), moderate/severe depression (β = −2.5; 95%CI −3.8, −1.2; p < 0.01), and longer time since HIV diagnosis (β = −1.2; 95%CI −2.1, −0.2; p = 0.01 for ≥10 years versus <5 years).

Sleep quality (PSQI)

Poor quality sleep (PSQI >5) was associated with HIV duration of ≥10 years (OR = 3.0; 95%CI 1.1, 7.9; p = 0.03), moderate/severe anxiety (OR = 5.8; 95%CI 1.2, 27.0; p = 0.03), mild depression (OR = 3.3; 95%CI 1.3, 8.4; p = 0.01), and moderate/severe depression (OR = 7.9; 95%CI 2.2, 28.4; p < 0.01). Prevalence of poor sleep quality was higher in those on EFV and INSTI-based regimens (48 and 51%, respectively) compared with those on non-nucleoside reverse transcriptase inhibitors (NNRTI) without EFV (39%); however, this demonstrated only a borderline association under multivariable analysis (p < 0.1 for EFV- and INSTI-based regimens).

Insomnia (ISS)

Insomnia (ISI ≥15) was associated with mild depression (OR = 4.7; 95%CI 1.0, 21.8; p = 0.05) and moderate/severe depression (OR = 16.1; 95%CI 2.6, 100.1; p < 0.01). Cannabis use and anxiety were associated with insomnia on univariable analysis (p < 0.05) but these associations were not held on multivariable analysis. Insomnia was more prevalent in those using EFV and INSTIs (28 and 27%, respectively) compared to the overall study population (22%); however, again this was not significant in multivariable analysis.

Sleepiness (ESS)

Daytime sleepiness (ESS ≥10) was associated with moderate/severe depression (OR = 4.6; 95%CI 1.2, 18.0; p < 0.01) and moderate/severe anxiety (OR = 4.4; 95%CI 1.0, 18.5; p = 0.05) in multivariable analysis.

Fatigue (FSS)

Fatigue was more likely in individuals experiencing mild depression (OR = 8.2; 95%CI 2.8–23.5; p < 0.01) and more so moderate/severe depression (OR = 18.4; 95%CI 4.9, 69.0; p < 0.01). A longer time since HIV diagnosis was also associated with fatigue (OR = 2.9; 95%CI 1.1, 7.7; p = 0.03).

Anxiety (GAD) and depression (PHQ)

In multivariable analysis, poor sleep quality (defined as PSQI >5) was associated with depression (OR = 8.3; 95%CI 3.1, 22.2; p < 0.01) and anxiety (OR = 13.2; 95%CI 3.6–48.0; p < 0.01). Depression was associated with use of crystal methamphetamine (OR = 4.2; 95%CI 1.1, 15.6; p = 0.03). Depression was more prevalent in INSTI users (25%) versus the overall population (18%); however, this was not significant after adjustment in multivariable regression. Older age (≥50 compared to <40 years, OR = 0.2; 95%CI 0.0, 1.0; p = 0.05) was protective against anxiety.

Wellness thermometer (wellness)

Lower wellness scores were associated with poor sleep quality as defined by PSQI (β = −1.4; 95%CI −1.9, −0.8; p < 0.01) and cannabis use (β = −0.7; 95%CI −1.4, −0.0; p = 0.04). Employment was associated with higher wellness scores (β = 1.0; 95%CI 0.2–1.8; p = 0.02) as was EFV-based ART (β = 1.0; 95%CI 0.1–2.0; p = 0.03 compared to non-EFV NNRTI regimens).

Discussion

In this study we found that prevalence of sleep disturbance in our cohort of PLHIV is higher than in general population. This finding is very relevant, as it is known that sleep disorders reported in PLHIV have been strongly associated with poorer health outcomes, cognitive impairment, and HIV-associated dementia.⁴ In line with our findings, a number of recent reports have provided contemporary evidence on the magnitude of the problem,¹⁹ but usually utilizing only single validated scoring assessments.

Our cross-sectional study provides evidence on the frequency and severity of sleep disorders in a contemporary cohort of PLHIV and has been done using validated multimodal questionnaires to investigate a number of sleep-related and mental health parameters.

Some studies report higher rates of mental health disorders in PLHIV than the general population and our results show sleep disturbance to be associated with mental health morbidity, notably anxiety and depression.^20,21

We observed several factors associated with reduced sleep quality, insomnia, and reduced wellbeing. Anxiety, depression, and longer HIV duration were associated with either poor quality of sleep or diurnal fatigue. Conversely, higher educational attainment was associated with better functional sleep outcomes. Methodologically, there was a high level of cross-validation between the instruments used to evaluate sleep, mood, and wellbeing.

The prevalence of substance use is reported to be higher among PLHIV than in the general population, ranging from 21 to 71%^20,22 and participants in our study self-reported higher rates of recreational drug use and ‘chemsex,’ than HIV-negative or undiagnosed MSM participants in the UK AURAH 2 study.²³ Recreational drug use in our study was associated with negative effects including anxiety and depression.

Furthermore, poor sleep quality, insomnia, and depression were more common in PLHIV on EFV and INSTIs, as previously described.^24–28

Sleep problems, both in the general population and in PLHIV, are multifactorial.²⁹ Among study participants, specific ART components and recreational drug use were found to impact sleep. The role of HIV immuno-virologic or lifestyle factors as contributing etiologies requires further investigation, but our data suggest that several avenues of investigation on lifestyle might be usefully pursued in subjects reporting poor sleep quality.

Our study presents some limitations. First, sleep disturbance was diagnosed based on self-reported questionnaire results rather than polysomnography data. Second, our cohort of PLHIV is homogeneous (mostly Caucasian MSM with adequately controlled HIV infection), so the evaluation of characteristics such as non-male sex and gender, other ethnicities, older age, and later or advanced HIV disease was limited. ART start dates were not recorded in our study and we would like to emphasize the importance of taking into account the lack of these data when interpreting our results given that recent ART change may interfere with sleep and daytime functioning. A final limitation of our study is a lack of specific data on alcohol consumption, as hazardous alcohol use has been shown to influence perceived sleep quality in PLHIV.³⁰ Our results are descriptive in nature and associations should not be assumed to be causative.

Nonetheless, our study has several strengths, given that we are able to describe different aspects of sleep and quality of life in a significant sample of PLHIV. The bidirectional impact of particular domains of sleep disturbance on psychological status and wellbeing in PLHIV has not been established thus far. Therefore, we have administered our own previously described³¹ and unique comprehensive multimodality assessments of sleep disturbance to better understand sleep outcomes in this well-defined inner-city cohort of PLHIV. The high level of correlation observed between the instruments used supports the use of this multimodality approach, in the formal assessments of changes in sleep, wellbeing, mood and quality of life in future studies, clinical practice, and to devise a model for person‐centered care.

In conclusion, results of our study show that sleep disturbance and mental ill-health remain prevalent in PLHIV and are associated with a series of modifiable behavioral factors. Improved screening and comprehensive management of recreational drug use, and psychological symptoms, including self-reported depression and anxiety should be the first steps toward creating strategies that can effect better sleep quality and thereby improve clinical outcomes.

Footnotes

Authors’ contribution

Conceptualization and manuscript preparation were done by all authors. Analysis was performed by BS.

Acknowledgments

These data were presented in part at the 17th European AIDS Conference (3–6 November 2019) Basel, Switzerland. We would like to acknowledge all the participants for their time and commitment. We thank all the 56 Dean Street staff who helped with the study.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Ana Milinkovic has received research and travel grants from Gilead Sciences, MSD, Janssen and ViiV Healthcare.

Suveer Singh has nothing to declare.

Bryony Simmons has nothing to declare.

Anton Pozniak has received support from ViiV Healthcare, Gilead Sciences, Merck Inc. and Janssen for research.

Marta Boffito has received travel and research grants from and has been advisor for Janssen, Roche, ViiV Healthcare, Bristol-Myers Squibb, MSD, Gilead Sciences, Mylan, Cipla, and Teva.

Nneka Nwokolo has received travel grants from and has been advisor for Janssen, ViiV Healthcare, and Gilead Sciences and since 2019 has been employed by ViiV Healthcare.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Ana Milinkovic

References

Gamaldo

Creighton

, et al. Evaluating sleep and cognition in HIV. J Acquir Immune Defic Syndr 2013; 63: 609–616.

Gamaldo

Spira

Hock

, et al. Sleep, function and HIV: a multi-method assessment. AIDS Behav 2013; 17: 2808–2815.

Chaponda

Aldhouse

Kroes

, et al. Systematic review of the prevalence of psychiatric illness and sleep disturbance as co-morbidities of HIV infection in the UK. Int J STD AIDS 2018; 29: 704–713.

Byun

Gay

Lee

KA.

Sleep, fatigue, and problems with cognitive function in adults living with HIV. J Assoc Nurses AIDS Care 2016; 27: 5–16.

Morphy

Dunn

Lewis

, et al. Epidemiology of insomnia: a longitudinal study in a UK population. Sleep 2007; 30: 274–280.

LeBlanc

Mérette

Savard

, et al. Incidence and risk factors of insomnia in a population-based sample. Sleep 2009; 32: 1027–1037.

Mahmood

Hammond

Nunez

, et al. Effects of sleep health on cognitive function in HIV+ and HIV− adults. J Int Neuropsychol Soc 2018; 24: 1038–1046.

Ibarra-Coronado

Pantaleon-Martinez

Velazquez-Moctezuma

, et al. The bidirectional relationship between sleep and immunity against infections. J Immunol Res 2015; 2015: 678164.

Liu

Chen

Roles of sleep deprivation in cardiovascular dysfunctions. Life Sci 2019; 219: 231–237.

10.

Institute of Medicine Committee on Sleep Medicine and Research. Extent and health consequences of chronic sleep loss and sleep disorders. In: HR

Colten

Altevogt

(eds) Sleep disorders and sleep deprivation: an unmet public health problem. Washington, DC: National Academies Press, 2006.

11.

Weaver

Laizner

Evans

, et al. An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep 1997; 20: 835–843.

12.

Buysse

Reynolds

CF 3rd

Monk

, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989; 28: 193–213.

13.

Bastien

Vallieres

Morin

CM.

Validation of the insomnia severity index as an outcome measure for insomnia research. Sleep Med 2001; 2: 297–307.

14.

Johns

MW.

A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep 1991; 14: 540–545.

15.

Krupp

LaRocca

Muir-Nash

, et al. The Fatigue Severity Scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol 1989; 46: 1121–1123.

16.

Spitzer

Kroenke

Williams

, et al. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 2006; 166: 1092–1097.

17.

Kroenke

Spitzer

Williams

JB.

The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001; 16: 606–613.

18.

Croston

Petrak

Ustianowski

Use of the wellness thermometer to improve consultations for patients with human immunodeficiency virus. Nurs Stand 2017; 31: 46.

19.

Gutierrez

Tedaldi

Armon

, et al. Sleep disturbances in HIV-infected patients associated with depression and high risk of obstructive sleep apnea. SAGE Open Med 2019; 7: 2050312119842268.

20.

Bhatia

Munjal

Prevalence of depression in people living with HIV/AIDS undergoing ART and factors associated with it. J Clin Diagn Res 2014; 8: Wc01–Wc04.

21.

Remien

Stirratt

Nguyen

, et al. Mental health and HIV/AIDS the need for an integrated response. AIDS 2019; 33: 1411–1420.

22.

Hartzler

Dombrowski

Crane

, et al. Prevalence and predictors of substance use disorders among HIV care enrollees in the United States. AIDS Behav 2017; 21: 1138–1148.

23.

Sewell

Cambiano

Speakman

, et al. Changes in chemsex and sexual behaviour over time, among a cohort of MSM in London and Brighton: findings from the AURAH2 study. Int J Drug Policy 2019; 68: 54–61.

24.

Gutiérrez

Navarro

Padilla

, et al. Prediction of neuropsychiatric adverse events associated with long-term efavirenz therapy, using plasma drug level monitoring. Clin Infect Dis 2005; 41: 1648–1653.

25.

Arenas-Pinto

Grund

Sharma

, et al. Risk of suicidal behaviour with use of efavirenz: results from the strategic timing of antiretroviral treatment trial. Clin Infect Dis 2018; 67: 420–429.

26.

de Boer

van den Berk

van Holten

, et al. Intolerance of dolutegravir-containing combination antiretroviral therapy regimens in real-life clinical practice. AIDS 2016; 30: 2831–2834.

27.

Hoffmann

Welz

Sabranski

, et al. Higher rates of neuropsychiatric adverse events leading to dolutegravir discontinuation in women and older patients. HIV Med 2017; 18: 56–63.

28.

Hoffmann

Llibre

JM.

Neuropsychiatric adverse events with dolutegravir and other integrase strand transfer inhibitors.

AIDS Rev 2019; 21: 4–10.

29.

Ramos Nogueira LF, Pellegrino P , Carvalho da Fonseca T, et al. Factors related to insomnia symptoms among people living with HIV. Rev Pre Infec e Saúde 2019; 5: 8481.

30.

Costa

Rojas Rojas

Laoste

, et al. Sleep disturbances in HIV-HCV coinfected patients: indications for clinical management in the HCV cure era (ANRS CO13 HEPAVIH Cohort). Eur J Gastroenterol Hepatol 2019; 31: 1508–1517.

31.

Elliot

Wang

Singh

, et al. Increased dolutegravir peak concentrations in people living with human immunodeficiency virus aged 60 and over, and analysis of sleep quality and cognition. Clin Infect Dis 2019; 68: 87–95.